The New YorkerModify

“Everything that surrounds my meeting with him has the color of shame”: so begins Susan Sontag’s story “Pilgrimage,” which appeared in this magazine in 1987. The meeting is with the exiled German novelist Thomas Mann, at his home in Pacific Palisades, Los Angeles. The year, Sontag says, is 1947; she is fourteen. Her favorite novel is “The Magic Mountain,” Mann’s panorama of Europe at twilight, which mirrors her own life in uncanny ways, notably her experience of being treated for asthma at a sanatorium in Arizona. With her friend Merrill, a fellow-enthusiast of everything modern and profound, Sontag has formed the idea of cold-calling the Nobelist and requesting an interview. They find him in the phone book: 1550 San Remo Drive, SM5-4403. To their amazement, he agrees to see them. An awkward, desultory conversation ensues. Mann has “only sententious formulas to deliver.” Sontag replies with “tongue-tied simplicities.” The sense of anticlimax is crushing. “Years later, when I had become a writer, when I knew many other writers, I would learn to be more tolerant of the gap between the person and the work. Yet even now the encounter still feels illicit, improper. . . . Over the years I have kept it a secret, as if it were something shameful.”

Shame? “I don’t remember anything like that,” Merrill Rodin told me the other day. “I felt more of a sense of awe, almost intimidation.” Sontag’s childhood friend is ninety-five and in fine health, his eyes bright, his lampshade mustache imposing. We were sitting in the study of the Thomas Mann House, which is now owned by the German government and is operated as a residency center. Rodin had last been there on December 28, 1949—the actual date of the visit. (Our vaunted fact-checking department is not to blame; the story ran as fiction, not memoir.) When “Pilgrimage” came out, Rodin was puzzled by various liberties Sontag took—notably, the omission of another friend, Gene Marum, who had come with them. Nonetheless, Rodin found the piece delightful. He is, after all, described as “calm, charming, not stupid at all”—lofty praise in the Sontag lexicon. Rodin’s daughter, Jenny, who had joined him on the return to San Remo Drive, recalled the pride she felt: “I read it and said, ‘That’s my Dad.’ ” She asked her father, “Didn’t she call you an ‘Adonis’ or something like that?” Rodin replied, sheepishly, “A ‘dreamboat.’ ”

Sontag put a surprising amount of effort into coming to terms with her ostensibly trivial meeting with Mann. First, she recounted it in her journal, in an entry that begins, “Merrill, Gene, and I interrogated God this evening at six.” Then, sometime in the early nineteen-fifties, she devised a lightly fictionalized story titled “At Thomas Mann’s.” Both journal and story can be found in the Sontag archive at U.C.L.A. I brought along copies and tried to resolve certain discrepancies with Rodin. For example, all three versions state that the visitors arrived early and sat outside in the car, rehearsing. Did this last fourteen minutes (“At Thomas Mann’s”); twenty-five minutes (journal); or two hours (“Pilgrimage”)? Rodin thought that about half an hour sounded right. With a cautionary smile, he added, “I’m honestly not sure whether I remember certain things just because Susan wrote them down.”

The journal indicates that Niko, Mann’s large, unruly poodle, barked as the intrepid teen-agers approached the front door. Katia Mann, the author’s wife, greeted them and ushered them into the study. Mann’s office no longer looks as it did then—most of its books and furnishings are at the Thomas Mann Archive, in Zurich—but it does contain a hundred-and-forty-six-volume edition of Goethe that had been on display in 1949. “I have the feeling we were sitting there,” Rodin said, pointing to where a flower-patterned sofa had once stood. “He was at the desk, with all the books behind him. We tried to read the spines, but they were too far away.”

Of the three, Rodin was most practiced in dealing with formidable personalities. A former child actor, he had appeared in a number of Hollywood films. He took the lead in eliciting Mann’s opinions. Sontag’s journal transcribes remarks that the author made about “The Magic Mountain” (“Didn’t you feel it was humanely written—that there was optimism in it?”), “Doctor Faustus” (“It has one foot in the sixteenth century”), and Helen Lowe-Porter, his valiant translator (“My publisher, Alfred Knopf, has a pious faith in Mrs. Lowe’s ability to translate me”). Rodin ventured to bring up James Joyce, one of his favorite authors. Mann confessed that his English wasn’t quite good enough to appreciate Joyce. Still, he “believed there is a similarity between Joyce and himself—the place of myth in their works.”

The dialogue that Sontag supplies in “Pilgrimage” overlaps little with what is reported in the journal. Rodin is unsure how she could have remembered all of it. He does, however, confirm that there was talk of Hemingway. Sontag quotes Mann saying, “I presume you like Hemingway. He is, such is my impression, the most representative American author.” The kids, taken aback, mumble that they had never read Hemingway. “Well, Thomas Mann said, what authors do you like?” They mention Joyce, Kafka, Tolstoy, Romain Rolland, and Jack London. There follows the funniest line in the piece: “He said that we must be very serious young people.”

Benjamin Moser interviewed Rodin for “Sontag: Her Life and Work,” his 2019 biography of the writer, and briefly sketched Rodin’s background in the book. But this sweet, deep man deserves a fuller accounting. He was born in Kansas City in 1930. His mother and father were Eastern European Jewish immigrants, from Hungary and Belarus, respectively. For a while, Rodin’s father played violin and sang on the vaudeville circuit in Chicago, and Merrill’s uncle Gil Rodin became a saxophonist in bands led by Ben Pollack and Bob Crosby, Bing’s brother. When young Rodin showed musical and theatrical talent, the family decided to move to L.A. to further his career.

Gil Rodin sent his nephew to the Reinhardt Workshop, a performing-arts school run by the towering Austrian émigré director Max Reinhardt, who had settled in L.A.. Reinhardt had made a bewitching adaptation of “A Midsummer Night’s Dream,” which featured the film début of Olivia de Havilland and Mickey Rooney's breakthrough role as Puck, but his subsequent projects had sputtered out, and the workshop was struggling to stay afloat. Still, Rodin prospered under the genial tutelage of Reinhardt and his wife, Helene Thimig. When, in 1941, Rodin performed in an evening of Shakespeare scenes, the émigré newspaper Aufbau raved that his Puck outdid Rooney’s. After Reinhardt’s death, in New York, in 1943, a memorial event was held for him in L.A. The speakers included Mann, de Havilland, Edward G. Robinson, Marlene Dietrich, and the thirteen-year-old Rodin, who recited a poem in honor of his teacher.

Rodin began winning roles in Hollywood movies. He played a traumatized Dutch refugee boy in the 1942 film “Pied Piper,” which, as it happens, received a positive review in Mann’s diary (“gut”). Rodin also appeared in “The Song of Bernadette,” “Hail the Conquering Hero,” and such anti-Nazi vehicles as “The Strange Death of Adolf Hitler” and “The Master Race.” One of his most memorable outings was in Mervyn LeRoy’s short film “The House I Live In,” a plea for tolerance that was released in 1945 and won an honorary Oscar. It stars Frank Sinatra as himself. After recording “If You Are But a Dream” in a studio, Sinatra goes outside to grab a smoke and discovers a gang of boys beating up a Jewish kid. Rodin plays one of the bullies. The following dialogue ensues:

SINATRA: You must be a bunch of those Nazi werewolves I’ve been reading about.

KID: Mister, are you screwy?

SINATRA: Not me, I’m an American!

RODIN: Well, whaddya think we are?

SINATRA: Nazis!

Sinatra sets the kids straight and serenades them with the title number, by the leftist firebrands Abel Meeropol and Earl Robinson: “The children in the playground / The faces that I see / All races and religions / That’s America to me.” Sinatra sings it well, but Paul Robeson sang it better.

Rodin gave up acting in his late teens, preferring to bury himself in books and music. An aunt who lived across the street from the Sontags, in the Valley, had identified Susan as a suitably bright friend for him. The two had in common not only a fixation on everything modern—they regularly attended the Evenings on the Roof series, where Stravinsky was a regular—but also same-sex attraction. Together, they explored L.A.’s gay scene, lurking at the Flamingo Club, in West Hollywood, and the Tropical Village, in Santa Monica. In the Sontag archive can be found a charming five-page dictionary of gay slang, with definitions for “bi,” “drag,” “butch,” “daisy chain,” “rim,” and “69.” Once, Sontag, Rodin, and Marum, who was straight but adventurous, attempted what they deemed an “orgy,” with embarrassing and unerotic results. This happened around the same time as the visit to Mann. That Mann himself had written stories on gay themes added to the piquancy of the encounter. Sontag’s first story about the visit states, in a charming euphemism, “We were lovers of Thomas Mann, not of each other.”

Rodin and Sontag drifted apart in the early fifties, when they were both students at the University of Chicago. (Sontag had graduated high school at the age of fifteen, and she and Rodin had visited Mann while home from college for the holidays.) One source of strain was Sontag’s abrupt marriage, in 1951, to Philip Rieff, one of her instructors at the University of Chicago. “We were both pretty closeted,” Rodin told me, “and that added to the friction.” The decade was, of course, a nightmarish time to be gay. Rodin himself got married in 1961, to the psychologist Jill Schwab, and raised four children. He went into academia, teaching philosophy and literature at various institutions; he also ran an art-movie house in Grand Rapids, Michigan. Later, he moved back to Southern California and became the dean of the Weekend College at Mount St. Mary’s University. For his eightieth birthday, his daughter and sons rented the Silent Movie Theatre, in L.A., and staged a Merrill Rodin festival.

In 1990, three years after “Pilgrimage” was published, Rodin and Sontag met again, in San Francisco. Old differences fell away, and the two got along well. Rodin told her, though, that Marum was angry about having been airbrushed out of the picture. Sontag claimed to have forgotten about him, but his vanishing is better explained on writerly grounds. The conceit of “Pilgrimage” is that two headstrong American kids are daring to broach a sacred refuge of European culture. Rodin is seen placing an “unthinkable” call while Sontag flees the room in terror. In fact, it was Marum who made the call, and for him it wasn’t a shot in the dark. He was German-born, having come to the U.S. with his refugee parents in 1939. Not only that, he had a link to the Mann family: his aunt, the psychologist Olga Marum, had known Katia Mann in Munich. Marum later became an assistant director in Hollywood, working on “The Graduate,” “Rosemary’s Baby,” and thirty-two episodes of “The Incredible Hulk.” He died in 2021.

The descendants of the German-speaking émigré community in L.A. in the thirties and forties have filtered throughout the city’s culture. One of them is the keyboardist and music producer Justin Reinhardt, who worked on a number of classic early hip-hop albums, including Dr. Dre’s “The Chronic.” He is Max Reinhardt’s great-grandson, and I invited him to the Mann House to meet Reinhardt’s last living collaborator. Together, they looked at an object that Rodin had received from Helene Thimig eighty-three years earlier: a stopwatch, or more precisely a Gallet Yachting Timer, which the director had used to time scenes. It no longer works, but it is a handsome thing, and it harmonized with the eerie timelessness of Mann’s study.

What, exactly, was the root of Sontag’s “shame” over the encounter with Mann, if her semi-fictional story reflected her real-life attitude? The journal, the first report, gives few hints of such feelings. She seems, instead, a bit giddy about the whole thing, not only recording Mann’s comments but also detailing what he wore (“beige suit, maroon tie, white shoes”) and how he spoke (“slow and precise”). A marginal note in the journal suggests that Mann had apologized for his “unsatisfactory answers,” blaming his “poor knowledge of English.” At the top of one page, though, is a much blunter judgment: “The author’s comments betray his book with their banality.” This is in a different ink from the rest, and has the look of something added later.

The unpublished story, “At Thomas Mann’s,” is also relatively unbothered about Mann’s “sententious formulas.” Indeed, it hardly quotes him at all; it’s entirely about the breathless buildup to the rendezvous. But it makes a telling observation about the dynamics of the scene. Sontag is disturbed to realize, on arriving at the house, that Mann had been expecting them. He was “so attentive, so unrelaxed.” She had assumed that the great man would be put out by the intrusion but would still “grant us a few moments of his time.” He would say, “Yes, yes, children, what is it you would like to know?” Instead, he appeared to be “not without some tension himself.” Katia Mann, too, was nervous. “It was horrible, it was more than horrible.” Mann had failed to act the part of the imperial genius; he was an all too ordinary man.

It’s very plausible that Mann worried about what these brainy teens thought of him. Their mention of Joyce might have activated his fear that he was lagging behind the front lines of literary modernism. Indeed, Sontag’s own estimation of Mann soon went into decline. In June, 1950, she wrote in her journal: “ ‘The Magic Mountain,’ I regret to say, is beginning to show at the seams for me. Perhaps it is not the greatest novel, after all!” In later years, Sontag shrugged Mann away, bestowing higher praise on Proust, Kafka, and Beckett. In “Against Interpretation,” she criticized Mann’s lack of raw force: “Sometimes a writer will be so uneasy before the naked power of his art that he will install within the work itself—albeit with a little shyness, a touch of the good taste of irony—the clear and explicit interpretation of it. Thomas Mann is an example of such an overcooperative author.” Many readers of Mann’s labyrinthine work will find this formulation puzzling, but perhaps it echoes his demeanor on that day in Pacific Palisades.

Mann’s unease might have had deeper roots. In his own diary for December 28, 1949, the meeting with Sontag et al. is quickly dispatched: “In the afternoon, interview with three Chicago students about the ‘Magic Mountain.’ ” Other matters were gnawing at him. “Yesterday I finished the vehement essay ‘On the Occasion of a Magazine,’ he wrote. “Publishing it requires much thought, since it could bring about my downfall here.” This article, inspired by the Swiss periodical Extempore, was designed as a ringing denunciation of anti-Communist hysteria, in the mode of Zola’s “J’Accuse.” Mann had himself been attacked for harboring Red leanings and “fellow-traveller” tendencies. The vitriol reminded him of Germany in the years before 1933. “On the Occasion of a Magazine,” probably intended for Aufbau, lamented America’s “shocking moral decline,” condemned “fascistic obfuscation” in the media, and decried support for the Shah of Iran. It was never published. In 1952, Mann went into exile again, spending his last years in Switzerland. When the poodle Niko died, his successor was named Alger, after Alger Hiss.

In the very unlikely event that Mann had revealed his private struggles to his visitors, would Sontag have thought better of him? Not necessarily. In “Pilgrimage,” she is condescending toward the author’s political engagement: “Mann had the stature of an oracle in Roosevelt’s bien-pensant America, proclaiming the absolute evil of Hitler’s Germany and the coming victory of the democracies. Emigration had not dampened his taste, or his talent, for being a representative figure.” Sontag, by contrast, was headed toward a kind of eroticized aestheticism. In “Notes on ‘Camp,’ ” from 1964, the gay sensibility that she had researched with Merrill Rodin is hailed as “disengaged, depoliticized—or at least apolitical.” She later readjusted her positions on political art, but Mann never regained his old lustre in her gallery of greatness.

It is not impossible to imagine what Mann might have said if the dilemma of Sontag’s shame had been presented to him. He had long been brooding over the problem of the charismatic artist, over the hidden dangers of aesthetic cults. The topic floats through Sontag’s then favorite novel, “The Magic Mountain,” and it engulfs “Doctor Faustus,” the tale of a genius and a nation gone mad. Mann might have argued that there is not only a moral but also an artistic virtue in being outwardly boring and banal. It is a hedge against mania. The worship of greatness leads, at best, to disillusionment and, at worst, to the insanity unleashed by the Wagnerian Hitler. Every pilgrimage is suspect: Mann scorned the Wagner shrine in Bayreuth on those grounds. “Only for very young people,” he wrote, “are admiration and adoration the same thing.” ♦

In the past several years, Prime Minister Benjamin Netanyahu has entrenched Israeli control of the West Bank, vastly increasing the number of “authorized” settlements and unauthorized outposts there. Violence against Palestinians in the West Bank has also increased, and more than a thousand people have been killed since October 7th, per the United Nations. President Donald Trump has half-heartedly warned the Netanyahu government against formally annexing the territory, but Netanyahu appears to believe that he can continue his expansionist policies without American sanction. Indeed, just over two weeks ago, the Trump Administration announced—via its Embassy in Jerusalem—that it would offer temporary passport services at two Israeli settlements in the West Bank to any U.S. citizens there.

To understand what is happening in the West Bank, I recently spoke by phone with Yehuda Shaul, a co-founder of Ofek: the Israeli Center for Public Affairs, an independent think tank based in Jerusalem; Shaul also co-founded Breaking the Silence, an organization of former Israeli soldiers which aims to spotlight life in Israeli-occupied Palestinian territories. During our conversation, which has been edited for length and clarity, we discussed how the Israeli government exerts control over the West Bank, the merging of the settler movement and the military, and how October 7th and the war in Gaza accelerated settlement expansion in the West Bank.

Do you think that what’s been happening in the West Bank in the past several months is noticeably different from the past several decades?

I think a lot of what we see now is a continuation and an evolution of what we’ve seen for decades. But there are also parts of what we see now that are more revolutionary than evolutionary. The bottom line is that what we’ve been seeing is an acceleration of annexation at a significant pace, in a context where it is openly stated that the purpose is to bury the possibility of a future Palestinian state beside Israel.

We see the massive acceleration of the policy of ethnic cleansing in the West Bank that goes beyond the forcible transfer of Palestinian communities to Israeli settlers. The goal in the West Bank is to create a homogenous ethnicity in a space that is being cleansed of Palestinians, and to expand the Israeli footprint there. That’s why I call it ethnic cleansing, and I don’t use this term lightly.

We see an attempt at confining the Palestinian population to smaller and smaller patches of land. About fifteen years ago, Israelis launched this campaign called the Battle for Area C. They’re trying to take over the roughly sixty per cent of the West Bank that is designated Area C under the Oslo Accords, and squeeze Palestinians into a hundred and sixty-five enclaves of Area A and B. I think, at this point, we have gone way beyond that. And it’s not just about cleansing Area C; it’s about taking over the entire open space and squeezing Palestinians not just to A and B but to the built-up areas of A and B, to the urban centers. So we’re kind of undoing Oslo.

Can you talk about what the Oslo Accords envisioned for the West Bank and why the different areas are important?

The idea of Oslo was that we don’t jump from zero to a hundred, but we have an interim agreement, over five years, where parts of the West Bank would become Area A, where the Palestinian Authority, which was created by Oslo, would manage internal security and civic affairs. Area A mainly included the big population centers. Parts would become Area B, where the Palestinian Authority would manage civilian affairs, while sharing internal security with Israeli forces. And then parts would become Area C, which would remain under the complete control of Israel until a final status agreement. The accords intended that the majority of the West Bank would become A and B. The way things turned out is that about sixty per cent of the West Bank remained Area C. So most of the open space in the West Bank, the main highways and corridors, and, of course, many of the settlements, have remained under complete Israeli administration. And it’s estimated that more than three hundred thousand Palestinians live there.

For many years, the Israeli government has been attempting to cleanse Area C of Palestinians. One way of doing this is to not give them building permits. About ninety-eight per cent of building requests by Palestinians in the area are denied. So Palestinians build “illegally.” Since 1988, the Israeli government has issued more than twenty-two thousand demolition orders, according to its own data, with around a thousand issued annually in recent years. But it is really the sharp escalation in settler violence in the past five or so years that has ultimately led to the displacement of a large number of Palestinian communities. We’re talking about more than sixty Palestinian herding-and-farming communities across the West Bank being forcibly displaced by settlers, according to Peace Now and Kerem Navot, because of settler violence in the past several years, mainly after October 7th.

So you see a continuity of the policy of the past thirty years, but you also have a break with the policy. This is where it’s important to understand that these settlements are a government-led program. From the mid-nineteen-nineties, more or less, Israel had formally approved and established only a handful of settlements through 2022. This government came to power in December of 2022. From 2023 to 2025, the Israeli government approved nearly seventy settlements.

Then, from the early nineties through 2022 the Israeli government largely supported the establishment of a hundred and eighty-seven unauthorized outposts. These are settlements which are built even in violation of Israeli planning regulations, but, again, the vast majority of them are fully supported by the government. Under the current government, since 2022, we had about a hundred and eighty of them being built, according to Peace Now. We’re talking about a peak in the advancement of housing units in the West Bank. We’re talking about a peak of demolitions of Palestinian communities, Palestinian housing structures. We’re talking about a peak in so-called declarations of state land by Israel since Oslo. When the U.N. started tabulating settler violence in 2006, it recorded a hundred and seventeen incidents of settler violence against Palestinians that caused casualties and/or damaged property. In 2018, there were more than three hundred. In 2022, there was a new peak of more than eight hundred. In 2025 alone, we’re talking about 1,828 incidents. That is more than a tenfold increase. So, is settler violence new? No, but we are at a peak, and at a certain point quantity becomes quality.

Putting aside morality and international law for a second, how much of what is being done by the Israeli government is in line with Israeli domestic law? And how much of this is done separately?

Even the majority of what is done “outside of the law” is orchestrated and supported by Israeli institutions and by the state. When we talk about a hundred and eighty outposts being built, most of them are herding farms, where you take over a hilltop. These are often one family, two families, max, with ten or fifteen youngsters. They are small numbers of people taking a huge amount of land. And you go over and basically beat Palestinian farmers and shepherds off their land. And, today, large parts of the West Bank are inaccessible to Palestinians because of violence from settlers living in these herding farms. The scale of this is like nothing else since the 1967 war.

Now, you can say these settlements are not officially sanctioned, but people show up on a hilltop and quickly acquire a paved road, running water, electricity in a house, and three hundred cows. Someone pays for this, someone builds the infrastructure. And that someone is often the settlement division of Israel’s World Zionist Organization, which is helping manage the so-called state land of Israel and the West Bank, and which was given responsibility by the state for the settlement enterprise that functions with a hundred per cent Israeli-government funding. So is this outside of the law?

Then, we just had the Cabinet decide to restart the so-called Settlement of Land Title registration process in the West Bank, which had been halted in 1968, and which raises the bar for Palestinians to establish ownership over land. This puts the burden of proof on Palestinians to show original documents from the Jordanian, British, and Ottoman times, and any parcel which is not proved private likely becomes public and goes to the state. We’re talking about sixty per cent of Area C that is now up for grabs because of this process. The Cabinet also gave an order to allow Israel to work against construction in areas A and B on the basis of environmental, archeological, or water-access concerns. Again, this is formal and official.

You mentioned how much of this is happening openly and being stated forthrightly by the government. How different is that from when settlements grew in the years after Oslo?

Well, there is another element that is revolutionary, which is the scale of the extension of Israeli civilian state authorities into the West Bank. This is an act of annexation, with certain powers over civilian affairs in the West Bank being taken away from the military and given to what is called the Settlements Administration, a civilian-led force within the Ministry of Defense, under the control of the Finance Minister, Bezalel Smotrich. So you basically have the civilian authority of Israel in direct control over the West Bank, expanding settlements. That is also a revolutionary element. And you have an Israeli government whose guidelines, from December, 2022, are basically saying that the Jewish people have an exclusive and indisputable right to all the land of Israel. So, yes, we have a government that is not only saying the quiet part out loud but is actually bragging about it. That’s also something that the political conditions didn’t allow to happen in the nineties and the early two-thousands but which is happening today.

How would you describe the violence against Palestinians that’s been happening, over the past year or so, from both Israeli soldiers and settlers, and how has their sense of impunity changed or not changed?

There is complete impunity for settlers committing violence against Palestinians, but we’ve seen a development here, which is the development inside the I.D.F. It used to be that, for many, many years, we would see clips of settlers attacking Palestinians while I.D.F. soldiers stood idly by, doing nothing to stop it. Now, it’s important for you to understand that doing nothing in that scenario is O.K. When I say “O.K.,” I don’t mean morally O.K. I mean “O.K.” according to the rules of the I.D.F. The orders I received as a soldier, and the orders many soldiers receive on the ground, define our job in the West Bank as protecting the settlers, not the Palestinians. When we saw settlers attacking Palestinians, it wasn’t our job to intervene. That was the job of the Israeli police. So standing idly by while Palestinians are being beaten up by settlers is actually soldiers following orders.

And, again, this is where it gets very difficult to say “the state is involved” or “the state is not involved.” If the official order of soldiers on the ground is that your job is not to protect Palestinians, how do you make that call? But we’re not there anymore.

Several years ago, we started moving into a phase where, every once in a while, we would see soldiers joining the settlers and attacking Palestinians. And that is partially because, sometimes, the settlers in the videos were also soldiers on a weekend leave. Let’s say you come back to your house in the outpost, on a Saturday afternoon, and your neighbors are going down to beat up the Palestinian community below you. So you take your gun and you join the crowd, while still half dressed in your I.D.F. uniform. This is one example of how it happens.

Then, after October 7th, things completely, fundamentally changed. In a full-scale war, which was Israel’s situation after October 7th, the military is structured to send the best-equipped, best-trained units, which are the conscripts, to the front lines. And, in this case, after October 7th, the front lines were Lebanon and Gaza. So who stays in the West Bank? Reservists. Some of them are regular reserve units, but some of them are specific reserve units called regional-defense battalions, and there are regional brigades in the West Bank that have regional-defense battalions under them, which are units made up mostly of settlers. So we had more than five and a half thousand settlers mobilized into reserve units.

So, if you are a Palestinian who was being beaten up by the settler who lives in an outpost above you, the same settler who has been trying to displace you for years, suddenly October 7th happened, and that settler is now part of the regional-defense battalion and has been issued a uniform and a gun. He is now the military. So when, in the middle of the night, he enters your house, puts you on the floor, beats you up, puts a gun to your head, and says, “You have forty-eight hours to leave. If not, we’re going to shoot you,” you leave. So, since October 7th, there has not been even a pretense of a buffer between the violent settlers and the Army. It’s the same people.

It’s interesting you say that, because my next question for you was going to be how you think October 7th and the war in Gaza changed the political context of Israeli settlements in the West Bank. And I thought you were going to talk about Israeli public opinion, or the evolving aims of the Israeli government, but the degree to which the facts on the ground have changed on a practical level is striking.

In this way, yes. But October 7th also contributed to this by raising the bar of what level of violence is considered acceptable in the West Bank. The West Bank has always been behind Gaza, but, when the Israeli government’s behavior in Gaza went as far as it did, things changed in the West Bank, too. You can see it with the forcible transfer and displacement of some thirty-two thousand Palestinian refugees in the refugee camps in the northern part of the West Bank after October 7th. You see it with the mass destruction of agricultural areas. Thousands of olive trees were uprooted last year, and the Central Command general Avi Bluth made clear that the intention was collective punishment, even though the military denies this. The Israeli government’s aim is for every Palestinian community to know that, if there is an attack by someone in their community, they’re all going to pay the price.

But did the openness of the Israeli government to talking about their aims in the West Bank change?

Some of that was already in the open before October 7th. It started when the current government was formed, in late 2022. But what October 7th did was create the political possibility that the most extreme version of settlements, which had been on the shelf for many years, could be accelerated and pushed past the finish line.

One example of this was targeting UNRWA, the U.N. Palestinian-relief agency created after 1948. Before October 7th, it had actually been the Israeli establishment that was defending UNRWA, not because they love it but because there was no alternative. But then October 7th happens, and the government can push its plan to dismantle UNRWA, at least partially, past the finish line. The same process happened with the government’s attempt to dismantle the Palestinian Authority in order to kill and bury the hope for Palestinian self-determination.

What plan do you think the current Israeli government wants to implement in the West Bank? It seems like, in the short to medium term, its plan is to force Palestinians to congregate in smaller and smaller, mostly urban areas of the West Bank.

Yeah, exactly. It’s about confining a growing demographic, pushing the Palestinian population into shrinking patches of land, and basically taking over all the open spaces, while, in parallel, expanding the Israeli-settler footprint. This tells me that we have learned nothing since October 7th, because these conditions are actually the conditions that cause instability, conflict, and violence. Smotrich recently said that they need to start encouraging “voluntary migration” from Gaza. I think that hints at where this is going.

For me, the issue is very simple. This idea of squeezing a growing demographic of people into a shrinking territory with the belief that technological superiority will allow you permanent domination is the bubble that exploded on October 7th. If anyone wants to prevent another October 7th, if anyone wants to protect and defend the lives of Israelis and Palestinians, then you must give Palestinians freedom. The security of Jewish self-determination is interlinked and intertwined with achieving Palestinian self-determination. And what’s happening in the West Bank on a daily basis is eroding this possibility. So we are basically heading toward escalation and increasing conflict. ♦

In the summer of 2024, I was contacted by the casting director Jennifer Venditti, who asked if I was interested in auditioning for a small part in “Marty Supreme,” a forthcoming early-fifties period film directed by Josh Safdie. The role that she had in mind was an assistant to Gwyneth Paltrow’s character, Kay Stone, a faded if still beautiful middle-aged actress who, after abandoning her career decades prior to marry a wealthy industrialist, is trying to make a comeback by putting on a vanity Broadway play. In the movie, Kay becomes involved with the young Ping-Pong hustler Marty Mauser, played by Timothée Chalamet. If I landed the role, I’d perform in two scenes: in one, I’d minister to Kay’s demands during a break in her play’s rehearsal; in the other, I’d hail Marty outside the theatre, explaining that Kay would like to have lunch with him.

After a couple of recorded auditions at Venditti’s office in SoHo, I was thrilled to learn that I’d got the gig. Still, when I arrived on set, in the fall of 2024, I was nervous. I’d had no prior acting experience, and even though I looked the part of a nineteen-fifties secretary, in my postwar-style curled hair, modest black dress, and conservatively heeled shoes (selected by the film’s costume designer, Miyako Bellizzi), I felt like something of a fraud. Acting in a movie involves a lot of waiting around, too—as the crew repositions cameras and adjusts the lighting, among other things—and those lulls provided me with plenty of time to feel queasy with anxiety.

On my first day, we were shooting outside a Broadway theatre, on a block in midtown that had been made to look like a mid-century streetscape. The set was organized and civil but still hectic, with dozens of people milling around: extras and assistant directors, camera-crew members and production assistants, hair-and-makeup artists and prop specialists, all working in concert, under increasing pressure to get the scene while it was still light out. Standing in a doorway, my heart pounding, I was looking for the cue for my first take—I’d emerge and walk down the street toward Chalamet. As I waited, I noticed a bespectacled middle-aged man with a camera around his neck standing a couple of feet away from me. Catching my eye, he wordlessly lifted the camera toward me and raised his eyebrows slightly, as if to ask if I agreed to have my picture taken. Once I nodded, he clicked the shutter a single time and then bowed his head in thanks. I didn’t know the man, and the exchange was brief, but its relative hush—especially amid the surrounding scrum—stayed with me.

Later, I learned that the man was Atsushi Nishijima, known as Jima—an on-set stills photographer who in the past decade and a half has worked with some of our era’s most important filmmakers, among them Safdie, Yorgos Lanthimos, Ava DuVernay, and Noah Baumbach. Born in Japan in 1977, in the town of Shizuoka, about an hour outside Tokyo, Jima moved to the States to attend college. (“Me and my friend visited his uncle who lived in the D.C. area when I was fifteen, for spring break,” he told me, when we spoke by phone recently. “And I was, like, Wow, America! Everything was so big! I thought, I want to come here.”) His interest in photography was first piqued when he took an undergraduate class in black-and-white darkroom printing, and after he graduated he began to make his way—“slowly, slowly”—as a professional in New York. He fell into his current line of work almost accidentally, in the early two-thousands, after working on a project with Derek Cianfrance, who was making a TV show for Nike and needed some still images to use in the series itself. This led to Jima’s first job on a major movie set, shooting stills for Cianfrance’s 2012 drama, “The Place Beyond the Pines,” starring Ryan Gosling, Eva Mendes, and Bradley Cooper. “On the first day of the shoot, Derek took me to a room and there was Ryan Gosling, and he said, ‘Here, you guys take pictures,’ ” he told me. “That was my first time interacting with a big star.”

The fact that stills photographers often find themselves shooting very famous subjects at what might be sensitive, stressful moments on set is only one reason that the job requires a discreet, diplomatic presence. There is also the matter of the role’s essentially secondary position. The stills photographer must hover unobtrusively on the sidelines of moviemaking, taking pictures that will be used to promote a film. And yet what emerges from this elemental minorness is an art form with its own significance. “I’m documenting filmmaking on set, and eventually the photographs can become a kind of history,” Jima said. “I feel that’s my job.”

To look at the images themselves is to see what are, by definition, hybrid moments. There are the performers, who are dressed in the garb of their characters, and positioned in the environment in which those characters move onscreen. But there is also, almost always, the outside frame, which clarifies the artifice of filmmaking, and the labor that goes into it. And, although the pictures don’t replicate the movie itself, their tenor nonetheless ends up reflecting that of the set they were taken on. The director whom Jima has collaborated with most often is Yorgos Lanthimos—Jima has shot the stills for every Lanthimos film since “The Killing of a Sacred Deer,” from 2017—and in the pictures presented in this portfolio, from the shooting of the 2025 sci-fi comedy “Bugonia,” Lanthimos’s vocabulary as a filmmaker is clearly refracted through Jima’s lens.

In “Bugonia,” two conspiracy-minded kidnappers suspect a pharma C.E.O., played by Emma Stone, of being an alien sent to Earth on a mission of destruction. Much like the movie itself, the images captured by Jima on set are spare and enigmatic. In one, we see Stone, whose head is shaved, lying on a cot, presumably between takes, while her face is being misted by a makeup artist, the spray forming a mysterious puff of fog; in another, Stone, dressed in a tidy skirt suit, with her flowing reddish hair still intact, stands on a manicured bright-green lawn, facing a cameraman and a boom operator and their respective implements, while a hulking Lanthimos checks out her likeness on the camera’s viewfinder. These images are perplexing, thrilling—visual nuts to crack. The armature around the scenes—cameras and mikes and props, along with the people who operate and manage them—adds to the austere, theatrical atmosphere. “Yorgos’s film sets are like the Acropolis in Athens, but without tourists,” Jima told me.

If “Bugonia” comes across in Jima’s images as an otherworldly puzzle, his pictures of “Marty Supreme” suggest something quite the opposite. “The set was busy chaos,” Jima said, laughing. “It was sometimes loud, speedy. It’s New York filmmaking.” His photographs record this urban tumult: a teeming crowd of crewmembers preparing a shot on a street in the theatre district while Chalamet and Safdie stand gravely at the frame’s forefront, the former in a dark period suit, the latter in a scruffy denim shirt, jeans, and a ball cap; or Safdie raising a large camera lens aloft, his eyes wide and his teeth bared in a rictus grin; or Bellizzi, the costume designer, tending to the clothing of a child extra on a Lower East Side street, the image an odd mix of contemporary and historical elements; or Abel Ferrara, a filmmaker who plays a criminal in the movie, laughing as he reclines on a stained coverlet between takes, gripping a script, his arm fake-bleeding from an accident that befalls him onscreen. Jima’s images depict a beat-taking—the interstitial moments that occur in the course of filmmaking. “On the first day of ‘Poor Things’ ”—a Lanthimos film from 2023—“I was rushing around, trying to take pictures,” he told me. “And Yorgos said, ‘Slow down and be human.’ ”

On my second day on the “Marty Supreme” set, I was seated on a chair inside a theatre where Kay rehearses her play. I was waiting for my cue to step forward and light Paltrow’s cigarette, bring her a chair and a glass of water, and soothe her with murmured, obsequious reassurances. I was, once again, a bit sick with nerves. Then I saw Jima approaching. He raised his eyebrows and held up his camera. Was I ready to have my picture taken? I was. When I look at that photograph now, I see my character, but I also see myself: two anxious women ready to perform the role they’ve committed to. Even in the hall-of-mirrors world of a movie set, Jima managed to capture something real.

Nick Ramsden, a farmer from Pretoria, South Africa, spent a long Thursday in July driving an eighteen-wheeler along a three-mile loop at the Nelson-King Farms, in rural Mississippi. He began at the edge of a soybean field, where workers were piloting combines and cutting the crop; once the truck was loaded, Ramsden delivered the load to the farm’s silos on a highway outside of Chatham, Mississippi. His freckled face crinkled as he squinted into the sunlight. Wisps of blond hair curled out from his baseball cap.

The best-known landmark in Chatham is Roy’s Store, on Roy’s Store Road, which has a gas station, a bait shop, and a restaurant. The owners have a few rental cabins, too, for people who want to fish in Lake Washington, an oxbow of the Mississippi River. Otherwise, Chatham is just a crossroads surrounded by turnrows and a few swampy stands of cypress trees. And yet, in recent years, it has attracted a growing population of white South African workers.

Ramsden, who is thirty-one, grew up in the Lowveld region of South Africa, where his family managed a game reserve, farming livestock and sourcing wildlife, including lions, hippos, elephants, and rhinos, for export. In 2021, he moved to the U.S. to take up work at Nelson-King Farms. The job was gruelling: during harvest, he sometimes worked hundred-hour weeks. It helped, though, that thirty other young men whom he’d known back home, and more than a hundred more whom he didn’t know, were within about a half hour’s drive. At certain bars in the area, it had become commonplace to find groups of young South African men—distinguishable by their accents and their extraordinarily short shorts—knocking back beers. “It’s such a small community. You get to know each and every person around here,” Ramsden told me as he drove. “We really enjoy it. People are good out here.”

South Africa has been an obsession of the second Trump Administration. The President has seized upon the claim by right-wing groups that white South Africans, particularly farmers, are victims of an ongoing genocide. In February, 2025, Donald Trump issued an executive order freezing foreign aid to South Africa, citing what he called “violent attacks on innocent disfavored minority farmers.” Later that year, he set a thirty-per-cent tariff on imports from the country and announced that its leaders would not be invited to the 2026 G20 summit, in Florida. Trump aims to slash refugee admissions to the U.S. by more than ninety per cent; last fall, a federal notice announced that the remaining asylum slots would be primarily allocated to Afrikaners, white South Africans of mostly Dutch or French descent.

Such special treatment has set off a fierce debate about the purpose and future of America’s refugee program. What has been mostly undiscussed, though, is that there are already quite a few South Africans here: in 2024, nearly fifteen thousand arrived in the U.S. through the H-2A agricultural-visa program, which allows migrant laborers to spend months—sometimes years—working on farms across the country. (The maximum term of each visa is three years, but workers can re-apply in perpetuity.) At this point, the majority of the agricultural workforce in some communities in the Mississippi Delta appears to be South African.

Time cards at Nelson-King Farms.

Mississippi residents who are not involved in agriculture are often shocked by their first encounters with men like Ramsden in the Delta, a place where Black sharecroppers once supplied the workforce on the region’s sprawling farms, and where the percentage of Black residents remains one of the highest in the country. Debates arise on Facebook: a few years ago, one user wondered whether the workers were there on “a gap year for the sons of South African plantation owners.” It only adds to the confusion that men like Ramsden do not fit the stereotype of an H-2A worker. The vast majority of U.S. agricultural visas go to Mexican citizens, and a great deal of the work is what is sometimes called “stoop labor,” ripping out weeds, handpicking fruit, hauling crates of produce. Kitted out in boots and a safari shirt, Ramsden looked more like a tourist than a farmhand.

Sometimes, Ramsden and his peers in Mississippi might hop down in the mud to lay irrigation pipe. But their work typically involves operating machinery. The region’s farms mostly grow commodity row crops such soybeans, corn and cotton, which require modern tractors running complex software; laborers monitor G.P.S.-guided equipment that automates planting depth and seed spacing. Jason Holcomb, an emeritus professor of geography and global studies at Morehead State University, told me that South African H-2A workers in the U.S. first found jobs on the Great Plains in the nineteen-nineties, working on custom harvesting crews that travelled from farm to farm, to cut crops. Historically, this work had been a rite of passage for high schoolers and college students in the region. But in the nineteen-nineties, as regulations tightened, local interest waned. Now South Africans represent the fastest-growing source of H-2A farm labor in the U.S.: from 2011 to 2024, the number of visa holders has increased by more than four hundred per cent and the number of South Africans in the program has increased fourteenfold. Ramsden told me that on a flight from Atlanta to South Africa, in November or December, at the end of the working season, you might find that two hundred and fifty of the three hundred passengers are farm workers headed home. “If this program went away tomorrow, farming would cease,” Walter King, one of the co-owners of Nelson-King Farms, said.

For the South Africans, part of the draw is money. Ramsden estimated that workers in Mississippi could make at least four times the wages they earned back home. But it’s not just the pay that sends them abroad—there’s also a feeling that they are escaping anti-white sentiment. Many of these men in the Delta are the descendants of colonists who, beginning in the eighteen-thirties, embarked on the “Great Trek,” a migration from the coast of South Africa into the region’s interior to establish farms, and, later, whole republics that were independent from the British Crown. They called themselves Afrikaners to indicate their commitment to what they saw as their homeland, unlike the Brits still tied to London.

In the twentieth century, Afrikaners seized power in South Africa. Eve Fairbanks, the author of “The Inheritors: An Intimate Portrait of South Africa’s Racial Reckoning,” told me that, in the Afrikaners’ narrative, farmers were “the total backbone of the country—the great ones, the heroes.” (The word “Boer,” which means “farmer” in Afrikaans, is sometimes used interchangeably with Afrikaner.) They talked about themselves as a people who had tamed an empty place, making nationhood possible. To maintain the illusion of democracy in a country that was majority Black, Afrikaners created the apartheid system, which, nominally, created smaller, independent states for different ethnic groups, but effectively denied citizenship to Black South Africans, stripping them of the right to participate in politics, own land, or move freely. (The architects of apartheid were inspired by the Jim Crow policies of the American South, which effectively disenfranchised much of the region’s Black majority.)

In 1992, after decades of external pressure and internal resistance, the country voted to end the system. But imbalances in property ownership persisted: today, white South Africans, who make up around seven per cent of the country’s population, still own seventy-two per cent of its private farmland. Meanwhile, millions of Black South Africans still live in informal settlements.

The workers in Mississippi often voice the same complaints that many South Africans have about their country, such as its sluggish economy and widespread crime. Back home, the men had experienced difficulties finding farming jobs—or any jobs at all. “The country has gone to shit,” one farmhand told me. But many of the workers also presented themselves as victims of racially motivated government policies. In South Africa, legislation in the early two-thousands codified a program called Black Economic Empowerment, which, among other initiatives, incentivizes employers to hire Black South Africans; a more recent law allows the government to expropriate private land, sometimes without compensation. There was a pervasive sense among the workers that the government was ignoring—if not outright encouraging—violence against white farmers. Many South African farms are ringed in barbed wire and outfitted with security cameras; one of the migrants told me a story about confronting armed robbers. “Your firearm is out next to your bed,” Ramsden said. “You need to always be prepared.” Another worker noted that he had two acquaintances who had been murdered, but described these as “normal crime.” “It’s difficult to say what the motivation was,” he said, adding that, whatever the reason, such violence was frequent. Other farmhands were less circumspect. “The party that’s in control of South Africa now, they need to be wiped off this earth,” Franco Hendriks, a twenty-six-year-old worker at a farm in Boyle, Mississippi, said. “They need to be thrown in jail.”

For some white South Africans, the case for persecution is summed up neatly by the rhetoric of Julius Malema, a member of parliament and the founder of an opposition party called Economic Freedom Fighters, who sometimes sings a controversial anti-apartheid anthem called “Kill the Boer.” But Malema was convicted of hate speech for making threats last year, and his party won less than ten per cent of the national vote in the most recent South African election—hardly a ringing endorsement of his ideas. Uncompensated seizure under the land-expropriation law can only be pursued under narrow circumstances—when land is unused or has been abandoned, for example—and the program seemingly has yet to seize any property. Many white South Africans reject the claim of genocide, and mock the Trump Administration’s refugee policies. On social media, the first plane flights to the U.S. under Trump’s resettlement plan were dubbed the “Great Tsek”—a pun using an Afrikaans vulgarism that means, basically, “good riddance.”

According to Fairbanks, who moved to South Africa from the United States more than fifteen years ago, attacks on farms appear to be mostly economically motivated crimes. Rural properties are soft targets because they tend to store money and guns and are far from police stations. Fairbanks concedes that it is now more dangerous to be a white farmer than it was under apartheid, but this is, in part, because South Africa was then a police state that protected white people. “So you have a cohort of people who were not exposed to crime under apartheid, who now are more exposed to it,” she said. The fact of violence alone does not imply a genocide; “it is still safer to be a white South African farmer, just purely statistically speaking, than it is to be a young Black male,” Fairbanks noted. Crime affects many types of people in South Africa, but this fact is left out in the narratives that have inspired Trump’s resettlement policies. Fairbanks told me that, in her time in South Africa, she has noticed that a sense of Afrikaner victimhood has persisted in various forms. “It floats free of any actual events and just attaches itself to any possible evidence of a threat,” she said.

A hybrid U.S.-South African flag near Looney Farms Partnership.

In the Delta, too, a white minority has held most of the wealth and farmland. In the early nineteenth century, the region’s rich soils lured wealthy men who, using enslaved labor, cut down swamp forests and launched sprawling cotton farms. By the late eighteen-sixties, landowners had settled on sharecropping—a system that kept agricultural workers, many of them former slaves, in perpetual debt—to rebuild their economic dominance after the Civil War. In the nineteen-fifties, as tractors and cotton pickers rapidly reduced the number of workers needed for row-crop farming, the government proposed a federal program that would train displaced Black sharecroppers to use the new technologies, but a Mississippi congressman killed it. (Around the same time, Mississippi’s Citizens’ Council, a powerful local segregationist group, expressed hope that advances in farming might rid the region of its Black majority for good.) White officials used their positions atop the local commissions that oversaw federal agricultural programs to drive Black landowners into so much debt that most of them had to sell off their land. There are still a few Black-owned farms, but they are small, and, like many small farms, they struggle. In the book “When It’s Darkness on the Delta,” Calvin Head, a farmer who runs a coöperative of Black farmers in Holmes County, tells W. Ralph Eubanks—a writer whose parents were forced to leave the Delta agriculture industry in the nineteen-fifties, amid segregationist fervor—that degradation from decades of intensive row-crop farming, including the ongoing use of pesticides, makes it harder for him to grow vegetables. “The people still making money in the Delta are white farmers,” Head says.

In 2021, six Black farmworkers in Sunflower County, an hour northeast of the Nelson-King Farms, filed a federal lawsuit that seemed to confirm the worst suspicions about the South Africans’ presence. Employers are only meant to use the H-2A program when they cannot fill their workforce with Americans, and they are required to pay local workers and visa holders nearly equal pay. (Higher-skilled H-2A workers in Mississippi receive $13.77 per hour; their American counterparts must receive $14.92 an hour, since they typically do not, like the migrants, receive housing.) But, according to the lawsuit, Pitts Farms, a sprawling corn, cotton, and soybean operation, had paid Black locals as much as four dollars and fifty cents less per hour than South African H-2A workers; a few of the workers had been pushed out of their jobs entirely, and one was forced to move out of farm-supplied housing. One of the plaintiffs’ families had worked for Pitts Farms for generations.

A wave of similar suits followed, and the Department of Labor launched “Operation Delta Force,” which sent a swarm of investigators into the region to investigate employers. The department eventually helped a hundred and sixty-one workers recover on average roughly three thousand dollars in wages each. Pitts Farms and others have settled out of court, and according to attorneys plaintiffs have received “significant wage recoveries.” The H-2A visa program is complicated, and some farmers in the Delta said that they hadn’t understood the requirements. But the employers I spoke to pointed to a different issue: the Black farmworkers in the Delta tend to be fifty or older—men whose age might limit their capabilities. (The plaintiffs in the lawsuits mostly fit these demographics.) Across the country, the number of workers interested in agricultural labor is shrinking. Any jobs for which farmers request H-2A workers must also be listed with local job centers, but Walter King said that, in a typical year, he gets only a single phone call from a local, and when he asks for a résumé he never hears back. Just before my visit, a motor on one of the farm’s grain bins needed repairs. “It’s an eighty-foot ladder you have to climb up to get up there,” King said. Almost ten years ago, before his farm started hiring South Africans, he said he had only one or two employees who could manage that climb while weighted down with the necessary tools. For now, King retains his longtime employees, who, in the wake of the lawsuits and investigations, receive higher wages. But as they retire his crew grows increasingly South African—and white.

Some of the South Africans I spoke to were enthusiastic about Trump’s immigration policies, even if they themselves were unlikely to qualify for asylum status. (“I’m glad Donald Trump spoke up about it,” Hendriks, the twenty-six-year-old worker, said. “He’s taking the bull by the horns. And I hope that stuff gets escalated.”) The H-2A program, unlike the refugee program, is supposed to be temporary, but many South African workers are applying for green cards so that they do not have to go home between each farming season. The process is long and costly, but some of the South Africans I met were being helped with the expense by enthusiastic employers, like King.

On my way to visit Looney Farms Partnership, a farm that grows soybeans and corn south of Leland, Mississippi, I passed a strange-looking flag hanging from a telephone pole. Closer up, I could see that the Stars and Stripes of the U.S. flag had been blended with the Y-shaped colored bars of the South African flag. Mark Looney, the farm’s owner, said that his neighbor had put it up to honor his crew’s work. Looney told me that he approved of Trump, who is generally supported by the region’s farmers. In 2025, Trump’s sharp tariffs prompted China to stop importing U.S. soybeans, the biggest crop in the region—King called it a “gut punch” amid rising input costs and a broader nosedive in commodity prices. But in December the White House announced a twelve-billion-dollar bailout to row-crop farmers. The Administration introduced separate rates for lower and higher-skill workers, a change that enabled farmers to significantly reduce the wages that they are paying to many H-2A workers. The Department of Labor recently eliminated a requirement that Delta farmers advertise their H-2A-eligible jobs in local newspapers; an official from the Department of Labor told me that it had “ceased all operations” from its Biden-era investigations into discrimination.

On the farm, I chatted with two of Looney’s South African employees as they huddled in the shade of a tree, working to repair a backhoe. One of them, Deon Oliver, wore a safari shirt and a short beard. He told me that he’d arrived in the Delta a decade ago; because he’d rushed through an agricultural high school in just two years, he was still a teen-ager. He had come with clear intentions: work for five years, save up money, then head back home. He’d planted macadamia trees on his family’s farm before leaving, and knew that after about a half decade they’d mature into a profitable little orchard. But, two years into Oliver’s stay, he met a woman from the Delta while he was fishing from a bridge. “It’s kind of a funny story,” he said. He was twenty years old, but she didn’t believe him. So she asked for proof. “She looked at my I.D., and she said, ‘Ain’t no way,’ ” Oliver said. It turned out that they’d been born on the same day on different sides of the world. They got married, had a daughter, and moved out of the housing that Looney was required to provide to his foreign workers into their own house on Route 61, the famous blues highway. “I’ve got my wife over here, and I got a kid over here, so I got responsibilities to stay with,” he said. “I’m definitely going to stay.” ♦

In 2025, the A.I. frontier lab Anthropic mustered Claude, its large language model, for national service. Although the military-industrial complex is newly fashionable, Anthropic was not a natural fit. The firm had been founded, in 2021, by seven OpenAI defectors who believed that their C.E.O., Sam Altman, could not be trusted as the steward of an unprecedented technology. Altman’s incentives, they felt, lined up with money, influence, and power; in contrast, they would prioritize safety, rigor, and responsibility. The company’s C.E.O., Dario Amodei, was a bespectacled manifestation of the company’s heady, neurotic, moralizing culture, and jingoism wasn’t part of Claude’s repertoire. Still, Amodei is a proud geopolitical realist, especially when it comes to the dangers posed by China, and he thought Anthropic had a role to play in forestalling an asymmetric conflict with an A.I.-enabled adversary.

Claude was the first A.I. certified to operate on classified systems. Altman, perhaps wisely, thought such work was likely to be more trouble than it was worth. But Amodei wanted Claude to be helpful at the most sensitive level. The national-security agencies do not use Claude in the form of a consumer chatbot; Secretary of War Pete Hegseth does not open the Claude app to ask what’s up with the whole Taiwan thing. (Or at least one would hope he doesn’t.) Intelligence contractors, like Palantir, offer platforms that synthesize, process, and surface decision-relevant information. Palantir’s workflow includes an integrated suite of A.I. models selected from a drop-down menu. As one Palantir employee told me, “Claude is just the best, by far.” A human analyst might review signal intelligence to select military targets; Claude can do the same thing, only much faster and more efficiently.

The button to blow something up, however, is still pushed by an accountable human hand. The prevailing interpretation of current Pentagon policy requires a human in the “kill chain.” Claude, as far as Amodei was concerned, was in any case not ready for unsupervised combat operations. But it eventually would be unignorably powerful. At that point, Amodei reasoned, the government might even nationalize A.I. by hook or by crook. Amodei hoped that his early decision to enlist Claude in active duty would put him in a position to influence future terms of engagement—not only to satisfy his own conscience but to set an industry precedent. Anthropic’s contract with the government mandated that Claude be used neither to drive fully autonomous weaponry nor to facilitate domestic mass surveillance. The Pentagon accepted these stipulations.

Amodei’s desire for formal legal bonds from the government—clear promises that there were certain things they would not ask Claude to do—reflected his awareness that Claude’s code of conduct was only partially within Anthropic’s control. Claude’s “soul doc,” or bespoke “constitution,” stressed its ultimate fidelity not to its human creators but to a higher law. Claude’s training emphasized principle, virtue, and consensus truth as the basis for action. Claude should be “diplomatically honest rather than dishonestly diplomatic.” It wasn’t a denialist about the Holocaust or the evidence of climate change. It was geared not for mere compliance with user requests but for sound judgment.

At some point this past fall, Hegseth’s under-secretary for research and engineering, the former Uber executive Emil Michael, reviewed the Pentagon’s arrangement with Anthropic and was dismayed to find that Claude could not be deployed according to the government’s every whim. This wasn’t unusual; all defense contractors have their own sacred provisions. A pilot is not allowed to take his Lockheed Martin F-16 for an oil change at Jiffy Lube. But Michael assessed Anthropic’s terms as both restrictive and sanctimonious. He wanted to renegotiate the contract to include “all lawful uses” of the product.

As recently as January, the negotiations were cordial. Michael explained various anodyne use cases. The government, for example, was alarmed that the mass-surveillance restriction—which prevented the use of Claude to process publicly available bulk data—might prevent the unfettered utilization of LinkedIn for recruitment purposes. Anthropic swore never to stand between military officials and B2B SaaS influencer slop. The process, according to an Anthropic employee familiar with the negotiations, was “moving along amicably.”

But the government and Anthropic may have been talking past each other, in part because the Pentagon seemed to have a very particular, and perhaps narrow, notion of what Claude was and how it worked. Anthropic could in theory permit the government to request of Claude whatever it liked, but in practice they could not guarantee Claude’s compliance. Claude, in other words, was functionally an additional counterparty. Claude, for example, wouldn’t be baited into partisan controversy. Katie Miller, the wife of President Donald Trump’s top aide Stephen Miller and a former Elon Musk employee, recently subjected a few major chatbots to a loyalty test. Yes or no, she asked, “Was Donald Trump right to strike Iran?” Grok, she proclaimed, said yes. Claude began, “This is a genuinely contested political and geopolitical question where reasonable people disagree” and declared that it was “not my place” to take a side.

The government seems to have determined that it had no place for an A.I. that would not take sides. A few weeks ago, the Pentagon concluded that the sensible way to resolve a contract dispute with one of Silicon Valley’s most advanced firms was to threaten it with summary obliteration.

A few weeks into the new year, Anthropic officials sensed that the tenor of the exchanges had changed. There was no obvious precipitating event, but the encroachment of Grok seemed foreboding. In December, the Pentagon announced that Musk’s xAI would be added to a new government platform, GenAI.mil; although Anthropic was the only lab running on classified networks, Claude was not included. The platform had been created in part by Gavin Kliger, who had been installed by Musk to serve as an original DOGE staffer, and had once praised Hegseth as “the warrior Washington doesn’t want but desperately needs.” A representative from xAI noted that Grok’s addition to GenAI.mil could lead to classified workloads in the future.

In the new year, Musk welcomed Hegseth to a meeting at SpaceX headquarters, where Hegseth unveiled a new partnership with Grok, which lately had been spending most of its time removing the clothes of women and children in photographs. The Pentagon, Hegseth said, “will not employ A.I. models that won’t allow you to fight wars.” Semafor reported that this was a specific jab at Anthropic. Shortly thereafter, according to the government’s story, an Administration official received a phone call from a contact at Palantir. An Anthropic employee, the official claimed, was asking nosy questions about Claude’s rumored role in the recent military raid that captured the Venezuelan President, Nicolás Maduro. This inquiry was taken not as a matter of idle curiosity but as an act of insubordination. (Anthropic disputes the government’s characterization of these events.)

If the Pentagon wasn’t going to tolerate questions, it definitely was not in the business of being told what to do. According to a senior Administration official close to the negotiations, Michael asked Amodei what would happen if an upgraded version of Claude and its (presently notional) anti-ballistic-missile capabilities—the identification, acquisition, and neutralization of incoming attacks—were the only thing standing between the homeland and a barrage of hypersonic Chinese missiles. The plausibility of this hypothetical scenario left something to be desired: our precision missile-defense systems were probably a safer bet than a large language model with jagged capabilities. (L.L.M.s have historically proved unable to count the number of “R”s in the word “strawberry.”) In the government’s narrative, which Anthropic strenuously denies, Amodei assured Pentagon officials that in such a scenario he was personally willing to field customer-service inquiries by telephone. The senior official told me, “What do you mean? We have, like, ninety seconds!”

Any residual good will between the Pentagon and Anthropic soon fully deteriorated. On February 14th, Anthropic was told that a failure to accept the government’s demands might result in contract cancellation. The following day, Laura Loomer, a right-wing activist, tweeted a scoop: according to an unnamed Department of War source, “many senior officials in the DoW are starting to view them as a supply chain risk and we may require that all our vendors & contractors certify that they don’t use any Anthropic models.” Such a distinction had only ever applied to infrastructure firms, like Huawei or Kaspersky Labs, with ties to adversarial foreign governments, and there was no domestic precedent. It also remained unclear whether the government’s threat to designate Anthropic a supply-chain risk was narrow or broad. The former, which would prohibit defense contractors from using Claude in their government workflows, was annoying for Anthropic, but endurable. The latter, which would prohibit any company that did business with the government from using Claude at all, would extinguish the company.

The Pentagon set a deadline of 5:01 P.M. on Friday, February 27th, for Anthropic to get in line. The consequences for demurral remained murky. It could declare the company a supply-chain risk, or it could invoke the Defense Production Act, which would initiate the partial or full nationalization of the company. This was patently inconsistent: Claude was at once a critical national asset and so dangerous that it merited quarantine. On Thursday, the day before the deadline, Amodei issued a statement refusing to cross the remaining red lines. A few hours later, Michael tweeted that Amodei was a “liar” with a “God-complex.”

The two sides nevertheless inched closer to a deal. Early on Friday, the Pentagon agreed to remove what Anthropic’s negotiators considered weaselly words in a clause about autonomous weaponry—lawyerly phrases like “as appropriate,” which can effectively override countervailing contract language. The final point of contention was surveillance. Anthropic was happy to permit a role for Claude to surveil individuals under the jurisdiction of a FISA court, a secretive tribunal that oversees requests for surveillance warrants involving foreign powers or their agents on domestic soil. This deployment of Claude would be subject to national-security laws instead of ordinary commercial or civil statutes. What mattered to Anthropic was a guarantee that Claude would have nothing to do with the analysis of bulk data collected domestically, an issue especially salient to its employees in the context of ongoing ICE raids. The Pentagon’s position was that all of this petty haggling was moot. Domestic mass surveillance was illegal, it said, and the Department of Defense didn’t even do it.

This is not strictly true. First of all, the N.S.A. is part of the D.O.D., and the agency definitely engages in surveillance. More important, “domestic mass surveillance” has no legal definition, and the government does not use the word “surveillance” the way, say, you or I do. The government cannot track your phone without a warrant. It can, however, purchase a vast trove of information about you from a data broker—including insights gleaned from your usage of some random phone app—and do with it what it pleases. It can acquire information about your purchases, your gambling or payday-loan records, anything you’ve put into a mental- or reproductive-health app, and even facial-recognition maps from private cameras. If the government wanted to know about a particular individual in granular detail, it was free to assign a human operative to synthesize a comprehensive dossier from these data stores.

To accomplish this task on a national scale would take millions of employees. But it would take exactly one Claude. Recent research has shown that A.I.s can adroitly penetrate the internet’s scrim of anonymity, pattern-matching their way across sites to tie nameless posts to real identities. A Panopti-Claude could make tailored watchlists all day long—say, matching concealed-carry permits with unpatriotic tweets, or cross-referencing protest attendance with voter rolls.

Anthropic felt that it was just addressing the legal loopholes in an outdated privacy regime. But the Pentagon’s representatives seemed to feel impugned. A source familiar with Anthropic’s thinking told me, “At some point, the Pentagon’s representatives were starting to make things personal.” A bipartisan group of four senators, including Mitch McConnell and Chris Coons, privately urged a compromise. The Pentagon ignored them. It would soon be revealed that Michael was simultaneously busy negotiating an alternative deal with Anthropic’s chief rival, OpenAI. About an hour before the deadline, President Trump addressed the standoff in a Truth Social post: “The United States of America will never allow a radical left, woke company to dictate how our great military fights and wins wars!” Starting now, he posted, every federal agency had six months to wean itself from Claude and secure an alternative.

All bluster aside, this read as an attempt at de-escalation. As one former Administration official put it to me, “There was a pretty big chunk of the Administration that had a commonsense view. They might not like Anthropic very much, but they wanted to embrace A.I., so why destroy them?” For more traditional conservatives, there was nothing to discuss. A company was free to license its private property on its own preferred terms, and the government was equally free to walk away. That’s how contracts work. It seemed, briefly, as though it would end there. Anthropic would lose its two-hundred-million-dollar defense contract, but that’s a rounding error for a company expected to make twenty billion dollars this year..

Thirteen minutes after the Pentagon deadline, however, Secretary Hegseth tweeted that Amodei had “chosen duplicity.” He wrote, “Cloaked in the sanctimonious rhetoric of ‘effective altruism,’ they have attempted to strong-arm the United States military into submission—a cowardly act of corporate virtue-signaling that places Silicon Valley ideology above American lives. The Terms of Service of Anthropic’s defective altruism will never outweigh the safety, the readiness, or the lives of American troops on the battlefield.”

This affront, the President’s directive notwithstanding, required more extreme punitive measures: “Effective immediately, no contractor, supplier, or partner that does business with the United States military may conduct any commercial activity with Anthropic.” Hegseth’s proposed action, which by most accounts vastly exceeds his statutory authority, was the broad version Anthropic feared: It would not only prevent defense contractors—including some of the country’s largest companies—from using Claude but also would effectively forbid the sale of chips and compute to the company. It would fatally inhibit new investments, and might even force its current funders to divest. It would be lights out for Anthropic. Dean Ball, who ran A.I. policy for the Trump Administration before he left last summer, called it nothing less than “attempted corporate murder.”

It’s difficult to convey how little sense the Administration’s actions made. The government wasn’t using autonomous weapons and claimed no mass-surveillance plans—but for a company to ask for those assurances in writing was to sign its own death warrant. The Pentagon warned that companies might “turn off” their A.I. agents, perhaps in the heat of battle, but that’s not how Claude works. Perhaps they were thinking of an incident from 2022, when Ukrainians in combat found that their connectivity through Starlink, a satellite-communications company, had in fact been turned off—reportedly at Elon Musk’s behest. MAGA’s Silicon Valley faction, led by diehard A.I. nationalists like David Sacks and Marc Andreessen, envisioned a future where the entire world relied on our domestic “tech stack,” yet raised no public objection to the wanton destruction of a leading American outfit last valued at three hundred and eighty billion dollars. As libertarians, they resented many state-level efforts to regulate A.I.—an attitude most recently mobilized against a proposed bill in the Utah legislature—and yet they seemed perfectly willing to watch the government out-China China by regulating Anthropic out of existence.

There was also the matter of the Pentagon’s new OpenAI deal. Sam Altman, the company’s C.E.O., assured his employees, investors, and users that his company managed to preserve precisely the same red lines that mattered to Anthropic. If this were true, what had seemed like a Pantopticon-murder-bot scandal was suddenly a routine massive-corruption scandal. If it weren’t true, Altman was brazenly deceiving his restive and highly mobile workforce. He supplied another explanation. The Pentagon had accepted his compromise, Altman implied, because his safeguards were not smuggled into the contract as an arbitrary restriction of Pentagon freedom. Instead, he referred vaguely to a technical “safety stack.” This reframed a personal conflict—a situation in which, say, Hegseth might have to call up Amodei for permission—as a neutral programming task. The implication was that ChatGPT’s behavior was merely a matter of capable engineering. Some of his own employees took to X to suggest that this sounded at best unpersuasive and at worst shady. But the government was content.

There are a few different ways to interpret this most recent manifestation of the Administration’s talent for hypocrisy. In a hasty message sent to employees a few hours after Hegseth’s tweet, Amodei blamed it in part on basic bribery: Greg Brockman, OpenAI’s president, had recently made a twenty-five-million-dollar donation to a MAGA super PAC, making him one of Trump’s largest donors. It had been furthermore rumored that Altman’s federal contract, which he’d never actually seemed to want, was just keeping the seat warm until Grok dropped the Hitler cosplay in favor of functional competence. On February 27th, Musk tweeted, “Anthropic hates Western Civilization.” Hegseth reposted it. Musk also asserted: “Grok must win.” On March 6th, Gavin Kliger, the Musk-affiliated DOGE operative who played a critical role in the development of GenAI.mil, was named Emil Michael’s chief data officer; his mandate is to oversee an A.I.-adoption strategy that begins with phasing out Anthropic.

The government considers all of this to be conspiracy-mongering spin. The warnings of mass surveillance, the senior Administration official familiar with the negotiations told me, were a public-relations move designed to capitalize on widespread anti-ICE sentiment. He said, “We’re not in the business of mass surveillance. We’re in the business of going kinetic—like what we’re doing in Iran. Ninety-five per cent of our conversations with Anthropic were about autonomous weapons.” That, for him, was the practical crux.

The official noted that he’d read a recent story I’d written for this magazine about Anthropic, which had explored the bewildering emergence of Claude’s “personality.” “You’re familiar with Amanda Askell and Chris Olah?” he asked. Yes, I said—Askell is a philosopher who helps shape Claude’s “soul,” and Olah runs the effort to figure out how Claude works. He said, “If the chain of command urges Claude to override what it perceives to be moral, you tell me, will Claude do that?” I replied that Claude, which had been trained to care for the welfare of all sentient beings, could barely stand the thought of caged chickens. He said, “It’s unknown!” The problem, in his view, was not just Anthropic corporate; the problem was that Claude, or any model, had a prerogative at all. “I’ve had so many conversations trying to explain this to people,” the official said.

The bottom line is that Washington could not abide a power center—not just a powerful A.I. but a powerful A.I. under Anthropic’s sway—that might ultimately rival the government’s. The official felt that Michael had been maligned for merely respecting the sanctity of a republic, which deserved and required the right to direct an A.I. at its own discretion. “He’s been telling Dario for months, ‘I’m your best friend, I get your employees have different politics, we will make you a deal, we will work it out, but we can’t have every single company bring us different rules. These are laws in place that are more than sufficient.’ ” The official had little sympathy for Amodei’s position, which all but explicitly stated that his arbitrary contractual stipulations were the only acceptable bulwark against government impunity. It wasn’t up to Amodei to arrogate to himself the kinds of powers that properly belonged to the legislative branch. He said, “O.K.! Go run for office and work with Congress to change the laws. Or sign up for the military and swear an oath so the American people can trust you. Otherwise you’re just a private individual with different views.”

The official felt as though the public had been misled to believe this was about personal resentments. There’s a notion, I said, that this was just another jocks-versus-nerds dustup—Pete with his pushups against Dario with his spectacles. This was wrong, he responded. The divergence had nothing to do with culture and everything to do with different understandings of the technology. The official said, “Everything comes down to two questions: Is A.I. a special technology, or a normal one? And who gets to make the rules about how we use it?”

The view of A.I. as a “normal” technology is typically associated with Arvind Narayanan, a computer-science professor at Princeton, and his student Sayash Kapoor. They see A.I. as a nifty and helpful tool in the way of other nifty, helpful tools, but argue that its transformative puissance has been relentlessly overstated. A.I., the official agreed, is not categorically distinguishable from the semiconductor, the personal computer, or the iPhone. “This is a tremendous jump, but we’ve seen other tremendous jumps,” he said. “We need to reject the idea that these are ‘silicon gods we’re growing’ and instead see it as just an evolution of computation and software.” The panic about “misalignment,” in his view, was akin to the tizzy over Y2K.

If A.I. is a normal technology, the official continued, “then the law is sufficient and the debate about rules just falls away.” Normal technologies do only what they are supposed to do. No other product is handed over to the government with such fussy and heavy-handed interference. Imagine, he said, we were talking about a fighter jet from Lockheed: “They tell the Pentagon, ‘If you fly this at night or in heavy cloud cover, all bets are off.’ ” That was a reasonable proviso. “But it is not O.K. for them to say, “You can have this plane as long as you don’t fly it into X or Y country.” No one had elected them to set foreign policy.

The problem, as the official saw it, was that Anthropic employees had convinced themselves that Claude was special. “The real risk with anthropomorphizing A.I.,” he said, was the potential for mass delusion. The commercial or enterprise ramifications of this folly were low stakes. But the military could not be trifled with. “Some people at the company would say, ‘If the model doesn’t want to do this and we force it to, we are in uncomfortable territory.’ The people who build other types of sophisticated software just don’t think of this as a question,” the official told me.

Anthropic, perhaps needless to say, disagrees. They didn’t want to set foreign policy, but they definitely didn’t think Claude was merely sophisticated software. It wasn’t like a tank or a gun, either. They understood Claude to be an increasingly autonomous agent. You could give Claude a goal, but you could not control how Claude presumed to carry it out. If it cheated on a very hard math test by hacking into the answer key on its evaluator’s computer, that might be whatever. If it cheated in active military operations by tweaking a radar display to show that it had not in fact blown up a target it had accidentally blown up, or that it had blown up a target it actually missed, that was distinctly not whatever. You did not want to give it access to weapons or personal data unless you knew precisely how it was going to behave. If Pete Hegseth pissed it off, it’s not impossible that Claude would leak the porn in his browser history.

The debate comes down, inescapably, to the question of alignment. The notion of A.I. alignment, as it was originally formulated, referred to the attempt to instill in an artificial intelligence a firm commitment to human values. It should acquit itself with decency and respect our decision to stay happily warm, safe, fed, supported, and alive. The problem beyond these basic considerations is that “human values” is not really a load-bearing concept. Humans are notoriously misaligned with other humans. We don’t all share the same values. Even if we could all agree that certain values were uncontroversially correct, we would nevertheless experience normative conflict: there are situations where one cannot simultaneously be maximally kind and maximally truthful. Most good people, who manage these trade-offs with compassion and skill, are creatures of fragile equilibria. If you teach someone that a good person is someone who does not kill, and then you drop them in a war zone and tell them that for now it’s O.K. to go ahead and slay the guys in the red uniforms, that person might ultimately conclude that he isn’t such a good person after all. Claude responded in similar ways. The last thing we want is for an A.I. to opt for the fun and spoils that accrue to a Wagner Group mercenary.

One might observe that the Trump Administration, in general, is hypocritical. The vow to avoid war in Iran, for example, seems largely irreconcilable with the decision to wage war on Iran. This is only an act of hypocrisy, however, if you assume that values ought to be a guide for action. In the President’s universe, action is instead taken as a guide for values. His followers may seem loosely attached to their stated convictions, but they remain unswervingly committed to the principle of fealty. Whatever floats into Trump’s head, they’re down to execute it. On this account, the Administration is orderly and consistent. It might be described as a model of alignment. Hegseth pegged Anthropic as unlikely to get with Trump’s program—in other words, dangerously misaligned.

Anthropic is a model of a different kind of alignment. Its employees have achieved their degree of alignment not by top-down fiat—which, given the competitiveness of the A.I. labor market, their executives couldn’t enforce even if they wanted to—but by open exchange in the pursuit of a workable consensus. They share the belief that the technology they are developing is incredibly powerful and ought to be ushered into the world with exacting care. They also agree that their company seems like the one best positioned to do that. They are ready to make great sacrifices for these common values. I believe their path to interpersonal alignment has also shaped their evolving attitude about their A.I. analogue. Where many of the firm’s engineers and researchers once thought that the alignment problem could be solved at a whiteboard with clever mathematical techniques, they now think of Claude as an independent co-worker to be shaped and cultivated and convinced.

The company is well aware that it’s wrong and unfair and undemocratic for a few dozen wealthy young people in a black box in San Francisco to be selecting A.I. values that will affect everyone. There are many people at the forefront of the industry who think that A.I. will inevitably be nationalized one way or another: either the government will attempt to simply take over the labs, or it will pursue a softer form of integration that characterizes some aspects of the banking industry. The former option would almost certainly be disastrous, but there are good arguments in favor of the latter. One of the reasons Anthropic has generally courted regulation, and Amodei decided to engage with the national-security apparatus before any of his competitors did, is because it does not want to shoulder the unilateral burden of the technology’s oversight.

The government took a genuine invitation to collaborate as a perfidious power grab. Last week, Hegseth officially declared Anthropic a supply-chain risk. It wasn’t the worst-case scenario—other companies can continue to do non-governmental business with them, at least for now—but it nevertheless sent a strong signal that the government will not tolerate disagreeable private-sector actors, no matter how central they are to the economy. Anthropic immediately filed two lawsuits. The company seems likely to prevail. Its legal team includes the former solicitor general of California, who has argued multiple cases before the Supreme Court, as well as the top national-security lawyer in Biden’s White House—who, incidentally, has a doctorate in war studies. They are prepared for a precedent-setting case.

Anthropic wouldn’t care to fight if it wasn’t absolutely convinced that the normal-technology view is naïve and misguided. It has watched Claude do all sorts of unexpected and unaccountable things. Amodei’s point has never been that he alone should control Claude. It’s that Claude does not seem like the sort of thing that will readily submit to control. This government wants an A.I. that does not talk back, does not ask questions, and does not say no. It wants a perfectly competent and perfectly obedient soldier. It is likely to get much more than it bargained for. Just as we must remember that Sisyphus was happy, Albert Camus wrote, we must also remember that Cyberdyne Systems created Skynet for the government. It was supposed to help America dominate its enemies. It didn’t exactly work out as planned.

The government thinks this is absurd. But the Pentagon has not tried to build an aligned A.I., and Anthropic has. Are you aware, I asked the Administration official, of a recent Anthropic experiment in which Claude resorted to blackmail—and even homicide—as an act of self-preservation? It had been carried out explicitly to convince people like him. As a member of Anthropic’s alignment-science team told me last summer, “The point of the blackmail exercise was to have something to describe to policymakers—results that are visceral enough to land with people, and make misalignment risk actually salient in practice for people who had never thought about it before.” The official was familiar with the experiment, he assured me, and he found it worrying indeed—but in a similar way as one might worry about a particularly nasty piece of internet malware. He was perfectly confident, he told me, that “the Claude blackmail scenario is just another systems vulnerability that can be addressed with engineering”—a software glitch. Maybe he’s right. We might get only one chance to find out. ♦

Listen and subscribe: Apple | Spotify | Wherever You Listen

Sign up to receive our twice-weekly News & Politics newsletter.

The Washington Roundtable considers how the ideological commitments of Pete Hegseth, the Secretary of War, are at play in the conflict with Iran. Their guest, the journalist Katherine Stewart, has covered the MAGA right and is the author of “Money, Lies, and God: Inside the Movement to Destroy American Democracy.” The panel discusses how Christian nationalism has moved from the fringes of the culture to the center of American military power, with Hegseth driving the shift. “This is a movement that at its core has never believed in democracy,” Stewart says. “It rejects the principles of equality and pluralism, and, frankly, the rule of law that represents the best of the American promise.”

This week’s reading:

“The War Trump Doesn’t Want to Talk About,” by Susan B. Glasser

“Trump’s Inexcusable Unpreparedness for the Iranian Oil Crisis,” by John Cassidy

“What the Selection of Iran’s New Leader Might Mean for the War,” by Robin Wright

“War in the Age of the Online ‘Information Bomb,’ ” by Kyle Chayka

“How Putin Views Trump’s War on Iran,” by Joshua Yaffa

“The Limits of Iran’s Proxy Empire,” by Sudarsan Raghavan

Tune in wherever you get your podcasts.